1: \begin{abstract}

2:   We consider non-linear evolution equations arising from mean-field

3:   limits of particle systems on discrete spaces. We investigate a

4:   notion of curvature bounds for these dynamics based on convexity of

5:   the free energy along interpolations in a discrete transportation

6:   distance related to the gradient flow structure of the

7:   dynamics. This notion extends the one for linear Markov chain

8:   dynamics studied in \cite{EM11}. We show that positive curvature

9:   bounds entail several functional inequalities controlling the

10:   convergence to equilibrium of the dynamics. We establish explicit

11:   curvature bounds for several examples of mean-field limits of various

12:   classical models from statistical mechanics.

13: \end{abstract}

14: